Visible and near-infrared driven Yb3+/Tm3+ co-doped InVO4 nanosheets for highly efficient photocatalytic applications

Kailian Zhang a, Jie Guan a, Ping Mu a, Kai Yang *ab, Yu Xie c, Xiaoxiao Li a, Laixi Zou *a, Weiya Huang a, Changlin Yu *d and Wenxin Dai b
aSchool of Chemistry and Chemical Engineering, Jiangxi University of Science and Technology, Ganzhou 341000, Jiangxi, China. E-mail: 19979706763@126.com; zoulaixi@jxust.edu.cn; Fax: +86-797-8312334; Tel: +86-797-8312334
bResearch Institute of Photocatalysis, State Key Laboratory of Photocatalysis on Energy and Environment, Fuzhou University, Fuzhou, 350002, China
cCollege of Environment and Chemical Engineering, Nanchang Hangkong University, Nanchang 330063, PR China
dSchool of Chemical Engineering, Key Laboratory of Petrochemical Pollution Process and Control, Guangdong Province, Guangdong University of Petrochemical Technology, Maoming 525000, Guangdong, China. E-mail: yuchanglinjx@163.com; Fax: +86-668-2982253; Tel: +86-668-2982253

Received 1st July 2020 , Accepted 19th August 2020

First published on 19th August 2020


To effectively enhance the utilization of clean sunlight energy, harvesting a large percentage of near infrared (NIR) light is significant. One of the commonly used effective methods for modifying semiconductors is by co-doping upconversion materials on semiconductors to heighten the photocatalytic efficiency. In this work, Yb3+/Tm3+ co-doped InVO4 nanosheets were synthesized by a facile hydrothermal path, and the crystal phases, morphologies, surface chemical compositions, as well as optical properties were characterized. Yb3+/Tm3+ co-doped InVO4 revealed significantly enhanced photoactivity towards chromium(VI) reduction and methyl orange oxidation under visible or NIR light irradiation. Furthermore, 5YT-IV presented the highest electrocatalytic performance and photocatalytic production of H2O2 under visible light irradiation, requiring low overpotential and low Tafel slope (390 mV dec−1) for hydrogen evolution reaction than that of the bare InVO4 (731 mV dec−1), and as well improved the yield of photocatalytic H2O2 production by about 3.5 times. This was primarily ascribed to intensive light absorption resulting from the benign upconversion energy transfer of Yb3+/Tm3+ and the boosted charge separation caused by the intermediate energy states. Moreover, the presence of h+ and ˙O2 as the main oxidative species played a significant role during the photocatalytic oxidation process of methyl orange and electrons played a decisive role in Cr(VI) reduction. This study provides a promising platform for efficiently utilizing the visible-NIR energy of sunlight in the field of photocatalytic H2O2 production and for alleviating environmental pollution in future.


1. Introduction

Over the past decades, increasing energy shortage and enormous environmental pollution have seriously restricted rapid economic and social development.1–4 For example, the hexavalent chromium (Cr(VI)), a major environmental pollutant, is applied in various fields, such as electroplating, paint fabrication, and metallurgical technology.5–8 In addition, due to the serious shortage of H2O2, many methods have been employed for abundant H2O2 production, such as anthraquinone autoxidation, oxidation of alcohols, and electrochemical synthesis,9,10 which require large amount of energy or organic solvents during the preparation process. Furthermore, they are prone to contamination by organic impurities when they are extracted from these systems.11 Fortunately, among the potential solutions, semiconductor photocatalysis, a promising and cost-effective technology in water or air purification, has attracted significant attention for utilizing solar energy in the clean energy production and photo-degradation of pollutants.12–14 Nowadays, the development of visible-near-infrared (Vis-NIR) light active photocatalysts is a significant aspect in utilizing the Vis-NIR light that accounts for 95% of the total solar spectrum.

So far, metal vanadate photocatalysts (MVO4 where M = Ag, Bi, In, etc.) as visible-light-driven objects have been under large-scale investigation and application due to their excellent light absorbing performance.15,16 Indium vanadate (InVO4), a significant visible-light responsive metal vanadate photocatalyst with a narrow band-gap of 2.0 eV,17–19 has attracted extensive attention owing to its excellent photocatalytic performance and high chemical stability. Nevertheless, it is not preferred due to the increased difficulty in separating the photoinduced free charge carriers. In a typical one-component system, electrons and holes generated on a single photocatalyst recombine so rapidly that their separation and migration under the action of the electric field is not feasible, and significantly inhibits the enhancement of the photocatalytic activity. Moreover, harvesting and utilizing, ultraviolet (UV) to the near-infrared (NIR) light with high-efficiency from the broad solar energy spectrum is still very challenging, in which 49% of the total solar energy reaches the earth's surface as the NIR light, 46% as the visible light and the remaining 5% is the UV light.20–22 Broad-spectrum active photocatalysts have become a promising alternative for efficiently increasing solar energy conversion from the UV to the NIR region. Several efforts have been made so to boost the utilization of UV and visible light but few efforts have concentrated upon the absorption efficiency of a larger fraction of NIR light.23 A limited number of measures have been adopted to investigate the utilization of NIR light. In order to rapidly harvest NIR photons and boost the NIR-induced photocatalytic performance, many scholars have been encouraged to integrate carbon quantum dots (CQDs) owing to their intense absorption of NIR light, or to connect the surface plasmon resonance (SPR) peaks of noble-metal plasmonic nanoparticles so as to respond to NIR light, due to which it is capable of red-shifting to the NIR region.24,25 But they are beyond practical applications due to photocorrosion susceptibility and high cost during processing. Hence, it should be a priority to develop a UV-Vis-NIR full-solar-spectrum responsive photocatalyst by designing novel efficient photo-energy conversion materials.

To date, doping the semiconductor's host lattice with transition metal ions has been considered as one of the most effective approaches to ameliorate the photocatalytic performance of broad light spectrum responsive photocatalysts,26 which can efficiently harvest NIR light and then emit visible and UV light through upconversion luminescence processes. The closely spaced energy levels presented in lanthanide ions are essential for accelerating photon absorption in succession and energy transfer under 980 nm NIR excitation. A bathochromic shift will be induced by transition metal ions, leading to a decrease in the band gap and the formation of a new intra-band gap state,27 which is helpful in efficiently boosting light absorption. Meanwhile, doping enables the photo-excited free charge carriers to easily separate and transfer with a shortened transfer distance from the interior to the surface,28 resulting in an enhancement in the photocatalytic performance. Hence, high-volume investigations have been devoted for modifying semiconductor materials with the application of lanthanide upconversion. Some related work has been done, such as YF3:Yb3+,Er3+/ZnS,29 Au/NaYF4:Yb,Er/WO3·0.33H2O-W18O49,30 and NaYF4:Yb3+,Er3+/Ag2CrO4.31 Numerous other photocatalysts, such as CdSe/TiO2/TiO2:Yb3+,Er3+,32 Yb/Tm co-doped In2S3,27 Nd/Er co-doped BiVO4,33 Yb/Tm co-doped BiPO4/BiVO4,34 Yb/Er/Tm co-doped BiVO4,35 and Er/Yb co-doped Bi2MoO6,36 have also presented excellent performance by designing the fluorescence resonance energy transfer. Yet, there have been few reports that have concentrated on directly doping lanthanides in the InVO4 semiconductor for enhancing the light absorption and separation of photoinduced free charge carriers.

In this study, a series of novel Yb/Tm co-doped InVO4 photocatalysts has been synthesized via a simple hydrothermal method and the photocatalytic performance towards chromium(VI) reduction, methyl orange oxidation, and photocatalytic H2O2 production was evaluated under visible and NIR illumination, respectively. The results show that the Yb/Tm co-doped InVO4 photocatalyst presented superior photocatalytic performance. Furthermore, the NIR to visible upconversion mechanism has been discussed. First-principles calculations based on density functional theory (DFT)37–40 were performed to investigate the band structure and electronic properties of Yb/Tm doped InVO4.

2. Experimental

All the chemicals used in the experiments were of analytical reagent grade and were used without further purification. Ytterbium nitrate pentahydrate (purity 99.99%) and thulium nitrate hexahydrate (purity 99.9%) were purchased from Aladdin Industrial Corporation. Ammonium metavanadate (purity 99.0%) was obtained from Sinopharm Chemical Reagent Co., Ltd (Shanghai, China). As for indium(III) nitrate tetrahydrate (purity 99.9%), it was purchased from Shanghai Macklin Biochemistry Technology Co., Ltd.

2.1 Sample fabrication

The synthetic method of InVO4:Yb3+,Tm3+ was performed using the following hydrothermal process. In a typical synthesis, 4 mmol of InCl3·4H2O and a different mole ratio of Yb(NO3)3·5H2O as well as Tm(NO3)3·5H2O were dissolved in 30 mL deionized water and stirred for 1 h. After that, 4 mmol of ammonium vanadate (NH4VO3) was introduced in the above mixed solution to form a yellow turbid solution. After stirring for another 30 min, the pH value was adjusted to 2 by nitric acid. The obtained resultant transparent suspension was transferred into a 100 mL Teflon-lined autoclave for heating at 180 °C for 12 h. After the autoclave was cooled down to room temperature, the collected slurry was centrifuged and washed successively with deionized water and absolute ethanol three times. Finally, the resulting product was dried in vacuum at 80 °C for 4 h. The synthesized products with different mole ratio of In3+[thin space (1/6-em)]:[thin space (1/6-em)]Yb3+[thin space (1/6-em)]:[thin space (1/6-em)]Tm3+ of 200[thin space (1/6-em)]:[thin space (1/6-em)]0[thin space (1/6-em)]:[thin space (1/6-em)]1, 200[thin space (1/6-em)]:[thin space (1/6-em)]4[thin space (1/6-em)]:[thin space (1/6-em)]1, 100[thin space (1/6-em)]:[thin space (1/6-em)]4[thin space (1/6-em)]:[thin space (1/6-em)]1, 50[thin space (1/6-em)]:[thin space (1/6-em)]4[thin space (1/6-em)]:[thin space (1/6-em)]1, and 25[thin space (1/6-em)]:[thin space (1/6-em)]4[thin space (1/6-em)]:[thin space (1/6-em)]1 were named as 0.5YT-IV, 2.5 YT-IV, 5 YT-IV, 10 YT-IV, and 20 YT-IV, respectively. Meanwhile, the pure InVO4 sample was synthesized with 4 mmol InCl3·4H2O and 4 mmol NH4VO3 in the absence of Yb3+ and Tm3+ for comparison under the same conditions, and it was denoted as IV.

2.2 Characterization

A Bruker AXS D8 Discover with Cu Kα (λ = 0.15418 nm) was provided to test the powder X-ray diffraction (XRD) patterns at a scan rate of 0.05° s−1 in order to characterize the crystal structures of the as-prepared samples. Scanning electron microscopy (SEM) was carried out on an FLA650F under an accelerating voltage of 15 kV, while high-resolution transmission electron microscopy (HRTEM) was conducted on a JEM-2100F instrument in order to record the morphology and structure of the obtained samples. X-ray photoelectron spectroscopy (XPS) was carried out to reveal the chemical states of the elements on a PHI Quantum 2000 XPS system, which was equipped with a monochromatic Al Kα source as well as a charge neutralizer. Fourier transform infrared (FT-IR) spectra were measured by a Nicolet 5700 FT-IR spectrometer. UV-Vis diffuse reflectance spectra (DRS) were carried out using a UV-Vis spectrophotometer (UV-2600, Shimadzu), which was calibrated with the BaSO4 reference standard. The upconversion photoluminescence (PL) spectra were conducted under 980 nm light excitation on an Edinburgh Instruments Model FLS980 spectrometer, coupled with an NIR laser, while the steady state photoluminescence spectra were obtained using a fluorescence spectrometer (FL/FS920).

2.3 Photoelectrochemical performance

Photoelectrochemical measurements were carried out in 0.5 M Na2SO4 electrolyte on a CHI660E workstation, in which the photocatalytic materials coated on FTO were used as the working electrode, and the saturated Ag/AgCl and platinum electrode were used as the reference electrode and the counter electrode, respectively. The linear sweep voltammograms (LSV) were operated under a scan rate of 10 mV s−1, irradiated by an Xe arc lamp (300 W) with λ ≥ 420 nm wavelength. For the preparation of the working electrodes, the FTO glass was washed twice with ethanol and deionized water, and dried at 80 °C. The slurry, made by mixing 10 mg of the sample with 0.5 mL of ethanol, was spread drop by drop onto the FTO glass with 0.25 cm2 area. Before the experiment could be carried out, the adhesion of the working electrode could be improved by drying at 50 °C for 2 h. The non-working area of the FTO conductive surface was painted with nail polish.

2.4 Photocatalytic performance

2.4.1 Photocatalytic production of H2O2. According to the previous method, the experiment of photocatalytic H2O2 production was operated under a 300 W xenon lamp with a 420 nm cutoff filter.31 20 mg of the sample was dispersed in the mixed solution with 18 mL of deionized water and 2 mL of absolute ethanol, which was afterwards transferred fleetly into the lining of a high-pressure reactor. Oxygen was injected subsequently for 10 min in order to exhaust air. After irradiation for 2 h under 0.2 Pa pressure oxygen atmosphere, 3–4 mL of the solution was collected and filtered by a syringe with a 0.45 μm filter membrane. 500 μL of the supernatant was added into the mixture consisting of 2 mL KI solution (0.1 M) and 50 μL ammonium paramolybdate (H24Mo7N6O24·4H2O, 0.01 M). Following this, the concentration of H2O2 production was estimate after ultrasonic dispersion for about 10 min. The absorption wavelength could be at about 352 nm, as measured by UV-Vis spectrophotometry.
2.4.2 Photocatalytic degradation of the pollutants. The performance for resolving contaminants over the synthesized materials was studied in a photochemical reactor, in which the photocatalytic system could be maintained at room temperature by circulating cool water. In a typical photocatalytic experiment, the suspension mixed with 30 mg of the photocatalyst for Cr(VI) reduction (or 15 mg photocatalysts for MO oxidation) and 50 mL of the pollutant solution (10 ppm chromium(VI), 10 ppm methyl orange) was magnetically stirred in the dark for 40 min for achieving the dynamic equilibrium of adsorption and desorption. Then, a 400 W metal halide lamp (light intensity: 148.36 mW cm−2) with a <420 nm UV filter and another >760 nm IR filter was applied for simulating visible light, and the NIR photocatalytic experiments were conducted by a 980 nm diode laser with a beam expander (VA-I-DC-980, VIASHO). After a given time interval, 4 mL of the mixed suspension was extracted for further centrifugation during the photocatalytic process, and was then analyzed using a UV-Vis spectrophotometer (UV-6300 Double Beam Spectrophotometer, Mapada Instruments). Moreover, the Cr(VI) concentration in the supernatant was measured by the 1,5-diphenylcarbazide colorimetric method and the degradation or reduction efficiency (η) could be calculated as follows
η = (1 − Ct/C0) × 100%
where C0 is the initial concentration of the pollutants and Ct is the temporal concentration.

2.5 Computational models and methodology

First-principles calculations based on density functional theory (DFT) were performed to investigate the band structure and electronic properties of Yb/Tm-doped InVO4. All the calculations were performed using the Vienna Ab initio Simulation Package (VASP) with the projector augment wave (PAW) method. The exchange and correlation potentials were described with the Perdew–Burke–Ernzerhof (PBE) generalized gradient approximation (GGA). The kinetic energy cutoff (420 eV) of the plane wave basis was used throughout and was well within the convergence of the total-energy calculation. The Brillouin zone was sampled with special k-points of a 2 × 2 × 3 grid for the relaxation process, whereas 20 k-points along high symmetry lines was used for the electronic structure calculations. The forces on each ion converged to less than 0.01 eV Å−1, and the stress was less than 0.02 GPa. All the atoms of the structures were fully relaxed to their equilibrium positions with an energy convergence of 1 × 10−5 eV and the atomic displacement was less than 5 × 10−4 Å, whereas the self-consistent field (SCF) tolerance was 5 × 10−7 eV. The pseudopotentials used for the adopted models were constructed by the electronic configurations as V-3d34s2, In-5s25p1, O-2s22p4, Yb-4f146s2, and Tm-4f136s2 for the ground-state electronic structure calculations. The convergences with respect to the cutoff energy and the k-points mesh were tested, and the results showed that the cutoff energy and the k-points mesh used in this work were enough for the system.

3. Results and discussion

Fig. 1 shows the XRD patterns of the as-prepared photocatalysts synthesized by hydrothermal synthesis. For pure InVO4, every diffraction peak can be clearly indexed to the orthorhombic InVO4 corresponding to the standard data (JCPDS no. 48-0898),41,42 which matched well with every diffraction peak of the as-synthesized products. Specifically, the diffraction peaks at 2θ = 18.59°, 20.84°, 23.02°, 24.89°, 31.07°, 33.05°, 35.21°, and 47.04° are distinctly indexed to the (110), (020), (111), (021), (200), (112), (130), and (222) crystal planes, respectively. The samples show a high degree of crystallinity and no peaks related to other impurities are observed for the 0.5YT-IV, 2.5 YT-IV, and 5 YT-IV samples, which indicate that doping with Yb3+/Tm3+ did not change the crystalline phase of orthorhombic InVO4 due to the low contents of Tm3+ and Yb3+. Importantly, two diffraction peaks at 25.5° and 34° correspond to YbVO4 (JCPDS no. 01-072-0271) or TmVO4 (JCPDS no. 01-072-0272) in the 10 YT-IV and 20 YT-IV samples because of the increasing concentration of Yb3+ and Tm3+ ions. A detailed comparison based on the corresponding enlarged patterns of the undoped and doped InVO4 samples presents the shifts in the diffraction peak position to lower 2θ angles, indicating the presence of an obvious change in the unit cell volume, due to which In3+ could be substituted by Tm3+ and Yb3+. It is well known that θ can vary with different d values according to the Bragg equation 2d[thin space (1/6-em)]sin[thin space (1/6-em)]θ = . Therefore, the peak at 25.5° shifts with different ion radii (RIn = 0.94 Å, RTm = 1.02 Å, and RYb = 1.008 Å). This result reveals that the crystal lattice of InVO4 can be incorporated and substituted by the Tm3+ and Yb3+ ions. In addition, it can be clearly observed that the increased doping content of Tm3+ and Yb3+ ions gradually weakens the intensity of the diffraction peak at 33°, suggesting that the sample prefers orientation growth along with the addition of Tm3+ and Yb3+ ions as the dopants. Moreover, the phase structure did not change for the other phases over the YT-IV sample, indicating the high crystallinity and successful preparation of the Yb3+/Tm3+ co-doped InVO4 samples.
image file: d0dt02318c-f1.tif
Fig. 1 XRD patterns of the InVO4 and Yb3+/Tm3+ co-doped InVO4 samples (a) and the corresponding enlarged patterns (b).

The morphologies of the as-obtained samples were investigated by SEM, TEM, and EDX mapping images. As illustrated in Fig. 2a, the typical SEM image shows that the obtained IV products consist mainly of high-yield sheet-like nanostructures with a floccule surface, without the presence of adhesive particles. After doping Yb3+/Tm3+ (Fig. 2b), it still preserves the nanosheet structure. In addition, similar morphological structures can be found from the SEM images of the 0.5YT-IV, 2.5 YT-IV, and 10 YT-IV samples (Fig. S1); however, excess Yb3+/Tm3+ doping (20 YT-IV) has a great structural effect on the host InVO4, changing its structure from sheet-like nanostructures to block aggregates. This can be attributed to the formation of REVO4 (RE = Yb3+, Tm3+), which is consistent with the conclusion from the XRD patterns. Fig. 2c and d clearly confirm the presence of thin nanostructures over the IV and 5YT-IV sheets, as can be seen in the TEM images. The HRTEM image of 5YT-IV (Fig. 2e) signifies a lattice spacing of ∼0.389 nm, which is consistent with the (111) plane of IV, thus revealing the high crystallinity of the 5YT-IV sample. Importantly, the EDX spectra (Fig. 2f), TEM image (Fig. 2g), and the corresponding EDX mapping images (Fig. 2h–l) of the 5YT-IV nanosheets confirm the presence of In, V, O, Yb, and Tm elements, thus indicating the successful preparation of Yb3+/Tm3+ co-doped InVO4.


image file: d0dt02318c-f2.tif
Fig. 2 The SEM images of IV (a) and 5YT-IV nanosheets (b); TEM image of IV (c) and 5YT-IV (d); HRTEM images (e) and EDX spectra (f) of 5YT-IV; TEM image of 5YT-IV (g) and the corresponding EDX mapping images of In (h), V (i), O (j), Yb (k), and Tm (l).

The elemental compositions of the as-prepared samples were probed by X-ray photoelectron spectroscopy (XPS). As presented in Fig. 3a, In 3d, V 2p, O 1s, Yb 4d, and Tm 4d are present in the XPS survey spectra of the 5YT-IV sample. The binding energies at 18.7 and 42.07 eV belong to O 2s and Yb 5p, respectively. The peaks corresponding to the V 3s and V 2s states occur at 69.8 and 632.98 eV, and the peaks at 120.7, 665.45, 704.93, and 827.79 eV are attributed to the presence of In 4s, In 3p3, In 3p1, and In 3s, respectively. In detail, the two individual peaks in the high-resolution In 3d spectrum (Fig. 3b) at 444.3 eV and 451.9 eV correspond to In 3d5/2 and In 3d3/2, respectively.43,44 Moreover, the peaks located at 516.4 eV and 523.7 eV in the V 2p spectrum are related to the V 2p3/2 and V 2p1/2 states that contribute to V5+ (Fig. 3c).45 The characteristic peak of lattice oxygen in InVO4 is located dominantly at 529.1 eV,46,47 and the shoulder peaks at 530 and 531.9 eV in Fig. 3d are predominantly assigned to the weakly bonded O and OH in the substance,43 respectively, which can further boost the trapping of photo-induced electrons and holes, thus resulting in the enhancement of the photocatalytic process. The symmetrical peaks of Yb 4d (Fig. 3e) centered at the binding energies of 197.5 eV and 199.1 eV correspond to the Yb 4d3/2 and Yb 4d5/2 states, in accordance with previous reports.48,49 The Tm 4d peak is positioned at 175.5 eV (Fig. 3f). In addition, a significant shift in the direction of low binding energies for In 3d, V 2p, and O 1s can be observed for the 5YT-IV sample (Fig. 3 and Fig. S2), suggesting that the presence of strong interaction between Yb3+/Tm3+ and InVO4. Based on the above detailed analysis, the XPS analysis further implies that Yb3+ and Tm3+ were successfully doped into the InVO4 materials.


image file: d0dt02318c-f3.tif
Fig. 3 XPS spectra of the 5YT-IV nanosheets: (a) survey, (b) In 3d, (c) V 2p, (d) O 1s, (e) Yb 4d, and (f) Tm 4d.

To further understand the surface chemical structure of the as-prepared IV and YT-IV samples, infrared spectroscopy was applied and is shown in Fig. 4. The FT-IR spectrum of pure InVO4 exhibits strong bands in the region of 1000–500 cm−1; the band positioned at 531 cm−1 is ascribed to the V–O–V vibration; on the other hand, the bands centered at 1052–634 cm−1 are assigned to the V–O–In stretching vibration.48,50 The peak at 773 cm−1 presents a significant change from a broad peak of the IV sample to a sharp peak for the doped InVO4 samples, which may be caused by Yb3+/Tm3+ doping. It is worth noting that the intensity of the peak at about 946 cm−1 is enhanced after the doping of Yb3+/Tm3+, which is attributed to the different dipole moment corresponding to the characteristic stretching of V–O–RE (RE = Yb, Tm) on YbVO4 and TmVO4.51 In addition, the bands located at 3155–3500 cm−1 correspond to the stretching vibrations of the surface hydroxyl groups. Furthermore, the band at 1622 cm−1 is attributed to the presence of residual hydroxyl groups. In case of the YT-IV samples, Yb3+/Tm3+ doping does not change the characteristic peaks, based on the comparison with that of the pure InVO4 sample.


image file: d0dt02318c-f4.tif
Fig. 4 FT-IR spectra of the IV and YT-IV samples.

Light absorption performance is known to be one of the critical parameters that can determine the performance of the photocatalyst. UV-Vis DRS of all the samples was conducted to further characterize the light absorbance properties. As depicted in Fig. 5a, bare InVO4 exhibits apparent fundamental adsorption in the UV-Vis region and an absorption edge at about 560 nm was observed, indicating the boosted adsorption towards visible light. Meanwhile, the light absorption intensity presents an apparent enhancement after the doping of Yb3+ and Tm3+ into the InVO4 crystal, indicating the possession of stronger visible light response for the YT-IV composites, which may be attributed to the decrease in the band gap and the boosted adsorption of photon energy from the upconversion process of Yb3+/Tm3+ in YT-IV. Furthermore, a distinct red shift exists in the YT-IV samples, which can be ascribed to the optical transition of the intermediate energy states within the energy gap of IV,27,50 which favors the utilization of largely solar energy and boosting of the photocatalytic performance. Generally, the band gap energies of the as-prepared photocatalysts are determined and listed in Table S1. An obvious change in the band gap energies can be obtained from the doped InVO4 samples. Co-doping Yb3+ and Tm3+ can lower the band gap for the 5YT-IV samples (1.95 eV) for the comparison of the bare InVO4 sample (2.07 eV), again indicating the formation of intermediate energy states and promotion of light absorption (Fig. 5b).


image file: d0dt02318c-f5.tif
Fig. 5 (a) DRS and (b) plots of (ahv)2versus photon energy (hv) for the samples.

As is known to all, the charge separation efficiency and transfer resistance are detrimental for the enhancement of the photocatalytic activity. This can be directly revealed by the transient photocurrent response and electrochemical impedance spectroscopy (EIS) on the electrode, for which the photoelectrochemical experiments were carried out to investigate the transmission and separation efficiency of the photogenerated charge carriers under irradiation of a 300 W Xe lamp with 420 nm cutoff tablets. As presented in Fig. 6a, based on the comparison with the IV sample, the transient photocurrent density on the YT-IV photocatalyst is remarkably higher, which favors the enhancement of the photocatalytic efficiency, thus implying even better light harvesting capability. The fast and steady photocurrent responses benefit from Yb3+/Tm3+ doping, which can boost the photo-induced generation and separation of free charge carriers.


image file: d0dt02318c-f6.tif
Fig. 6 Photoelectrochemical measurement over the pure IV and YT-IV samples: (a) photocurrent measurements; (b) electrochemical impedance spectra (inset: the equivalent circuit mode).

In general, the radius of the EIS arc for the photocatalyst can also directly uncover the resistance in photo-induced charge transfer, for which the small EIS arc radius reflects the low transfer and separation resistance in this system. Obviously, the YT-IV photocatalyst exhibits a lower EIS arc radius compared with that of the pure IV sample (Fig. 6b); in particular, the 5YT-IV sample presents the smallest one, further indicating more efficient photoinduced charge transfer capability and enhanced separation efficiency. Combined with the above analysis, it can be deduced that doping Yb3+/Tm3+ in the IV sample could be anticipated to boost the photocatalytic performance.

Interfacial charge transfer kinetics was used to get a better insight into the equivalent circuit model (the inset of Fig. 6b) consisting of Warburg element (Zw), constant phase element (CPE), electrolyte resistance (Rs), and charge transfer resistance at the electrolyte–electrode interface (Rct). The values of Rct are 17.66, 17.3, 16.25, 15.09, 16.41, and 17 Ω cm2 for IV, 0.5YT-IV, 2.5 YT-IV, 5 YT-IV, 10 YT-IV, and 20YT-IV, respectively, where a smaller Rct value implies a much smaller interfacial resistance between the electrolyte and the electrode. Therefore, doping Yb3+/Tm3+ on the IV sample can enhance the photocatalytic performance due to the much smaller interfacial charge transfer resistance. Moreover, it is well known that the Fermi level (EF) is closer to the flat band potential from the Mott–Schottky relationship of the n-type semiconductor. Therefore, the energy-level structure of the undoped IV and 5YT-IV samples can be gained by Mott–Schottky analysis, in which there exists a significant negative shift from −0.98 V (vs. Ag/AgCl) for IV to −1.07 V (vs. Ag/AgCl) for 5YT-IV (Fig. S3). The conduction band edge potentials of undoped IV and 5YT-IV are 0.358 eV and 0.448 eV, respectively.

Upconversion luminescence spectra of the obtained YT-IV samples can be observed under the excitation of 980 nm NIR light. The result, as demonstrated in Fig. 7a, makes it clear that the emission peaks at 474 nm (blue light), 621 nm (red light), and 800 nm could be matched well with the 1G43H6, 1G43F6, and 3H43H6 transitions of Tm3+,27,35,52,53 respectively. Yb3+ can be excited efficiently by 980 nm wavelength, where it possesses a large absorption, and then successively transfers the energy to Tm3+, thus signifying the formation of the upconversion process. Also, 5YT-IV does not present the most obvious transformation in the visible light region, which may be because the energy from Yb3+ to Tm3+ is further transferred to the InVO4 semiconductor under 980 nm NIR irradiation.54 Therefore, doping Yb3+/Tm3+ can extend the semiconductor's absorption to NIR light and can effectively harvest the NIR photons by InVO4.


image file: d0dt02318c-f7.tif
Fig. 7 The upconversion emission spectra upon 980 nm excitation (a); schematic diagram based on the energy-level structures and upconversion processes of the Yb3+ and Tm3+ ions (b); the CIE coordinates of all the samples (c).

Inspired by the NIR photons, the Yb3+/Tm3+ co-doped InVO4 semiconductor could adsorb the lower energy photons, followed by conversion into the emitted photons with higher energy via upconversion. The mechanism of upconversion processes can be illustrated by the Yb3+/Tm3+ energy level structures that are presented in Fig. 7b. Upon NIR irradiation, Yb3+ ions can be excited by 980 nm pump photons from 2F7/2 to 2F5/2, and then the energy can be transferred to the Tm3+ ions by phonon-assisted energy transfer, resulting in the population of the higher energy levels of 3H5, 3F3, and 1G4 over Tm3+ through successive energy transfers. The non-radiative relaxation of 3H53F4 and 3F33H4 populates the 3F4 and 3H4 levels of the Tm3+ ions.55 The next energy transfer to the Tm3+ ion finally populates the 1G4 state. Tm3+ ions of the 1G4 energy levels are more inclined to relax non-radiatively to 3H4 owing to multiphonon relaxation rather than radiative decay to 3H6. Hence, the radiative transitions of 1G43H6 and 1G43F4 deliver weak blue and red emission positioned at 474 nm and 621 nm, respectively. Fig. 7c shows the color coordinates of the fluorescence based on the CIE chromaticity diagram, and the upconversion spectra of all the samples are distributed in the white light region, corresponding to this result presented in Fig. 7a. Therefore, the InVO4 semiconductor can further absorb and utilize these fluorescent excitation energies, resulting in enhanced light adsorption and boosted photocatalytic performance for photocatalytic Cr(VI) reduction, MO oxidation, and production of H2O2 through irradiative energy transfer as well as non-irradiative energy transfer processes.56–58

Water electrolysis is usually used to obtain pure hydrogen, which is a convenient and clean way, and a high current density at low overpotential can be achieved over an efficient hydrogen evolution reaction (HER) catalyst.59,60 Meanwhile, to better comprehend the enhancement in interfacial charge transfer, HER performances were studied to investigate the changes in the active sites and interfacial charge transfer by Yb3+/Tm3+ doping. The electrocatalytic HER performances over the bare IV and 5YT-IV samples were estimated in acidic solution (0.5 M H2SO4) in a three-electrode configuration. As shown in Fig. 8a, the 5YT-IV sample reveals excellent HER performance, while pure IV presents no HER activity as the 5YT-IV sample requires lower overpotential for electrocatalytic HER under the same current density than that required by pure IV. The HER kinetics were tested by the corresponding Tafel plots, as show in Fig. 8b. 5YT-IV presents a lower Tafel slope (390 mV dec−1) than that of pure IV (731 mV dec−1), revealing that 5YT-IV delivered highly efficient interfacial charge transfer and more significant enhancement performance for HER. Double-layer capacitance (Cdl) was determined to assess the effective electrochemical active area. Obviously, as shown in Fig. 9, the Cdl of 5YT-IV (30.67 μF cm−2) is much larger than that of bare IV (7.77 μF cm−2), suggesting that Yb3+/Tm3+ doping could induce an enhancement in the active area and is helpful for HER performances as it provides more exposed sites.


image file: d0dt02318c-f8.tif
Fig. 8 LSV curves with a san rate of 10 mV s−1 (a) and the corresponding Tafel plots (b) of bare IV and 5YT-IV samples.

image file: d0dt02318c-f9.tif
Fig. 9 Cyclic voltammetric curves of IV (a) and 5YT-IV (b) at different scanning rates; plots of the capacitive current measured at −0.35 V vs. RHE as a function of the scan rate (c).

The activity of photocatalytic H2O2 production was evaluated under xenon lamp illumination with a <420 nm UV filter. As described in Fig. 10, H2O2 could be generated over the IV and 5YT-IV systems, while no H2O2 production could be observed in the absence of the photocatalyst and the light source, unambiguously indicating that the generation of H2O2 is controlled by photocatalysis. Interestingly, the yield of H2O2 over the YT-IV suspension increased slowly and then decreased with incremental doping amounts of Yb3+/Tm3+, which showed a stronger photocatalytic activity for H2O2 production compared with that of pure IV, thus indicating more positive action for the photoreduction of O2 to H2O2. The enhanced photocatalytic activity for H2O2 production maybe because of the boosted visible light absorption, as confirmed by the DRS result, and the improved efficiency towards photoinduced interfacial charge transfer and separation.


image file: d0dt02318c-f10.tif
Fig. 10 Photocatalytic H2O2 production over different photocatalysts.

The photocatalytic activities for chromium(VI) reduction and methyl orange oxidation (MO) were studied under visible light or NIR light irradiation, as shown in Fig. 11 and 12 over the as-synthesized YT-IV and pure IV photocatalysts. As shown in Fig. 11, the reduction efficiencies of Cr(VI) over different YT-IV photocatalysts exhibited excellent photocatalytic performance in comparison to pure IV under visible or NIR light irradiation, suggesting that doping Yb3+ and Tm3+ in InVO4 is essential for enhancing the photocatalytic performance. As shown in Fig. 11a, it is worth noting that no signal can be traced for the removal of Cr(VI) in the absence of the photocatalyst, indicating that it is stable enough that it cannot be directly reduced by light. In addition, it could be clearly concluded that co-doping Yb3+/Tm3+ remarkably enhances the photocatalytic performance, following the order of 5YT-IV > 2.5 YT-IV > 10 YT-IV > 20 YT-IV > 0.5 YT-IV > IV. Meanwhile, 5YT-IV exhibits optimum photocatalytic performance for the removal of Cr(VI). The same result with enhanced photocatalytic performance in the presence of YT-IV can also be concluded under NIR light irradiation (Fig. 11b). As presented in Fig. 11c, the comparison of the photocatalytic performance of Cr(VI) reduction under visible or NIR irradiation clearly proves the contribution of Yb3+/Tm3+ doping, in which their optimum photocatalytic performances are 21.03% and 37.45% over the 5YT-IV photocatalyst (only 3.38% and 6.22% for the IV nanosheets) under NIR and visible light irradiation, respectively. The rate constants are shown in Fig. S5, which were estimated to be 0.00298 min−1 under visible light irradiation and 0.00156 min−1 under NIR irradiation. They are higher than that of the bare IV sample (0.00042 min−1 and 0.00024 min−1). In addition, the 5YT-IV photocatalyst also presents boosted photocatalytic performance for 5 ppm Cr(VI) reduction under visible light irradiation (Fig. S4a), reducing 42% of Cr(VI) in 150 min. But the degradation efficiency is only 16% for the undoped InVO4 samples. The 5YT-IV photocatalyst can reduce almost 99.3% Cr(VI) (10 ppm) under 500 min irradiation, while it was only 25% for the bare IV sample (Fig. S4c). This is ascribed to Yb3+/Tm3+ that convert long wavelength light (such as NIR, visible) into short wavelength light (visible, UV), which can again stimulate InVO4 for the production of photoexcited charge carriers.


image file: d0dt02318c-f11.tif
Fig. 11 The photocatalytic reduction performance of Cr(VI) over the as-synthesized YT-IV photocatalysts under visible (a) and NIR (b) irradiation; the comparison of reduction performance (c).

image file: d0dt02318c-f12.tif
Fig. 12 The photocatalytic performance for MO degradation over the as-synthesized YT-IV photocatalysts under visible (a) and NIR (b) light irradiation; the comparison of MO degradation performance (c).

The photodegradation of MO in aqueous solution was also implemented to prove the contribution of Yb3+/Tm3+ doping to the photocatalytic activity of InVO4 under visible or NIR illumination. As introduced in Fig. 12, there is no photocatalytic activity in the absence of the photocatalyst, implying that no photolysis occurs but photocatalysis plays a key role in MO degradation. Comparatively, the YT-IV nanosheets present far superior photocatalytic ability towards MO degradation than pure IV, as shown in Fig. 12a and b. The 5YT-IV photocatalyst shows degradation rates of 23.1% and 37.7% (low photocatalytic efficiency of 1.4% and 2.9% for pure IV) under NIR and visible light irradiation, respectively (Fig. 12c). The rate constants were calculated to be 0.00307 min−1 under visible light irradiation and 0.00172 min−1 under NIR irradiation, while it was only 0.00018 min−1 and 0.00011 min−1 for the undoped InVO4 sample (Fig. S6), respectively. In addition, the 5YT-IV photocatalyst can degrade 56% MO (5 ppm) under visible light irradiation (Fig. S4b), while it is only 15% for undoped InVO4. Moreover, as can be seen from Fig. S4d, the time required by the best sample to degrade MO to at least >95% is ca. 350 min. In addition, the total organic carbon (TOC) of MO over the 5YT-IV sample has been tested (Fig. S4e); it could be found that the TOC amount decreases as the illumination time increases, suggesting that MO is decomposed rather than decolorized. With an increasing ratio of Yb3+/Tm3+ doping on InVO4, the photocatalytic efficiency of YT-IV presents variable tendency of first being enhanced and then being reduced, which can be attributed to the light decay caused by the Yb3+ loadings.61 Inspired comprehensively by the above investigations, the boosted photocatalytic efficiency maybe due to the boosted NIR absorption, the improved upconversion, higher efficiency of interfacial charge transfer, as well as electron/hole pair separation generated by intermediate energy states in YT-IV. In addition, the overall view of the comparison of the NIR photoactivities towards dye degradation over other Yb3+/Tm3+ upconverting photocatalysts is displayed in Table S2, wherein it can be seen that Yb3+/Tm3+ co-doped InVO4 delivers a relatively high photodegradation activity under NIR light irradiation, indicating that the Yb3+/Tm3+ co-doped InVO4 photocatalyst presents better promising applications as an NIR-activated photocatalyst.

Moreover, the stability of the better photocatalyst 5YT-IV was studied by the recycling test during MO degradation. As shown in Fig. S7, with an overall view of the MO degradation performance, a slight change in the degradation activity is presented after three successive cycles. It indicates that YT-IV photocatalysts present good stability as well as better promising application. In addition, the XRD pattern of the 5YT-IV photocatalyst after three cycles for photocatalytic MO degradation presents no evident variation in the crystalline structure (Fig. S8), thus further confirming the chemical stability of Yb3+/Tm3+ doped on InVO4.

To further investigate the recombination capability of the photogenerated charge carriers, steady-state photoluminescence (PL) spectroscopy was performed for the as-synthesized samples under 380 nm wavelength excitation (Fig. 13). The stronger PL intensity suggests higher recombination capability of the photoexcited charge carriers,62 which will inhibit the enhancement of the photocatalytic performance. It can be clearly observed that the 5YT-IV sample exhibits a dramatically lower PL spectra than that of the IV photocatalyst, implying that it presents impeded recombination and promoted separation efficiency of the charge carriers over the 5YT-IV sample, which is in accordance with the results of transient photocurrent response and EIS.


image file: d0dt02318c-f13.tif
Fig. 13 The steady-state PL spectra with 380 nm excitation wavelength.

In order to identify the enhanced photocatalytic mechanism in depth, the primary reactive species were recorded by the trapping experiment for MO degradation and Cr(VI) reduction over 5YT-IV under visible light irradiation. During the photocatalytic reaction process, 1 mM tert-butyl alcohol (t-BuOH), 1 mM p-benzoquinone (p-BZQ), and 1 mM EDTA-2Na as the scavengers were mainly selected to capture the hydroxyl radical (˙OH), superoxide radical (˙O2), and hole (h+), respectively. In addition, 1 mM KBrO3 was used to capture the electrons during Cr(VI) reduction. As introduced in Fig. 14a, the MO degradation efficacy in the presence of the 5YT-IV photocatalyst weakens dramatically from 37.7% to 14.8% and 13.2% with the respective addition of EDTA-2Na and p-BZQ, respectively, implying that the presence of h+ and ˙O2 as the main oxidative species caused significant degradation. On the contrary, the addition of t-BuOH has a slightly negative effect on the degradation, denoting that ˙OH did not participate in the reaction as the predominant active species. As displayed in Fig. 14b, the photocatalytic effect of Cr(VI) reduction is inhibited by the addition of KBrO3 as an electron scavenger, while it is promoted by the addition of EDTA-2Na as an electron donor, suggesting that electrons played a decisive role in Cr(VI) reduction.


image file: d0dt02318c-f14.tif
Fig. 14 Trapping experiment towards the active species in MO degradation (a) and Cr(VI) reduction (b) under visible light irradiation.

To understand the boosted photocatalysis over the 5YT-IV sample, as shown in Fig. 15, the supercells with 4 × 2 × 3 unit cells containing 576 atoms were constructed for Yb/Tm-doped InVO4. In the supercell, four In atoms were replaced by the Yb dopant and one In atom was replaced by Tm in the doping case, yielding a dopant concentration of 4.2% and 1.0% for Yb and Tm, respectively. As tabulated in Table 1, the fully-relaxed lattice parameters of InVO4 and Yb/Tm-doped InVO4 crystals are listed. It can be seen that the calculated lattice parameter is 5.844 × 8.677 × 6.754 after the geometric optimization of InVO4, which is very similar to the literature value of 5.769 × 8.546 × 6.565.63 It shows that the model structure is consistent with the reality and the calculation process is reasonable.


image file: d0dt02318c-f15.tif
Fig. 15 Models of pure InVO4 and Yb/Tm-doped InVO4: (a) pure InVO4 unit cell; (b) Yb/Tm-doped InVO4 super cell.
Table 1 Comparison of lattice parameters (Å)
System Lattice parameters (Å)
a b C
InVO4 calculation value 5.844 8.677 6.754
InVO4 literature value63 5.769 8.546 6.565
Yb/Tm doped InVO4 4 × 2 × 3 super cell 23.378 17.355 20.263


The introduction of impurity Yb and Tm atoms causes the lattice to be distorted, and the volume is slightly greater than that of the pure InVO4 4 × 2 × 3 super cell. The reason for this phenomenon is mainly because the ionic radius of In3+ is 0.94 Å, which is close to that of Tm3+ (1.02 Å) and Yb3+ (1.008 Å).

To evaluate the thermodynamic stability of the Yb and Tm dopants in InVO4, the formation energies (Ef) were calculated via the following formula:

 
Ef = EdopedEInVO4EYbETm + EIn(1)
where Edoped and EInVO4 correspond to the total energies of the pure and Yb/Tm-doped InVO4, respectively. EYb, ETm, and EIn are the chemical potentials of Yb, Tm, and In metals, respectively. According to this definition, the nanomaterials with negative Ef are energetically favorable and can be fabricated easily with more intimate interface binding. The obtained result of −5.02 eV of Ef for Yb/Tm-doped InVO4 indicates a stable structure.

To have an insight about the electronic distribution properties of Yb/Tm-doped InVO4, the electron density differences around the Tm and Yb atoms can be seen from Fig. 16. Herein, the red regions indicate electron accumulation and the blue regions indicate electron depletion. Obviously, there is a considerable charge accumulation close to the V and O atoms, indicating electron transfer from Yb or Tm atoms to the O or V atoms. The distribution of the charge density is nearly spherical around the O atoms and a small overlap in the charge density exists between the Tm (Yb) and O atoms, thus presenting an ionic feature. Meanwhile, In atoms have small charge accumulation, indicating the covalent bonding of In and O atoms.


image file: d0dt02318c-f16.tif
Fig. 16 Charge density difference map of Yb/Tm-doped InVO4.

The change in the band gap due to the doping of Yb/Tm can be demonstrated by the density of states (DOS). It clearly shows that InVO4 has a direct energy gap of 2.923 eV (Fig. 17). The calculated band gap is different from the value of the experiments due to the difference between the ideal crystal cell and the actual crystal. But it can transform our results of the corresponding electronic band structures. The energy gap of Yb/Tm-doped InVO4 is 2.292 eV, which significantly decreases compared with that of pure InVO4. A decrease in the band gap leads to a much wider visible-light absorption range. This is consistent with the results of the DRS experiments.


image file: d0dt02318c-f17.tif
Fig. 17 The projected band structures of IV (a) and 5YT-IV (b).

Inspired by the above investigations, as shown in Scheme 1, a possible enhancement mechanism is proposed over the Yb3+/Tm3+ co-doped InVO4 photocatalyst. In detail, under the irradiation of NIR light, the electrons over the Yb3+ ions could transit from the 2F7/2 level (ground state) to the 2F5/2 level (the excited state), which will then be introduced into the higher energy levels (3H5, 3F3, and 1G4) of the Tm3+ ions through the energy transfer process. After absorbing the energy transferred from upconversion luminescence, the InVO4 semiconductor can photo-induce the formation of free charge carriers over the CB and VB, and the photogenerated electrons could be separated immediately in the presence of intermediate energy states, leading to lower recombination and highly efficient separation rate. Cr(VI) could be reduced to form Cr(III) owing to the more negative CB edge potential of YT-IV than the Cr(VI)/Cr(III) potential of +0.51 V vs. NHE. Furthermore, the O2 of the reaction system can contact easily with the photoexcited electrons as the CB edge potential of YT-IV (−0.448 eV vs. NHE) is far higher than the standard redox potential Eθ (O2/˙O2) of −0.33 eV vs. NHE, resulting in the generation of ˙O2 radicals for H2O2 production or MO oxidation.


image file: d0dt02318c-s1.tif
Scheme 1 Schematic illustration of the photocatalytic process under Vis-NIR light irradiation.

4. Conclusions

In summary, the boosted photoactive Yb3+/Tm3+ co-doped InVO4 was successfully synthesized via the hydrothermal process. The upconversion process induced by Yb3+/Tm3+ offered a certain amount of suitable energy in this system for the activation and subsequent photocatalysis under NIR irradiation. The excellent photocatalytic performance for photocatalytic H2O2 production, Cr(VI) reduction, as well as MO oxidation and electrocatalytic HER performance was benefited by the shortened transfer distance of the free charge carriers owing to the generation of intermediate energy states that are induced by the doping of Yb3+/Tm3+, enhanced light absorption, and reduced chance of recombination of the free charge carriers in the body. Hence, it highlights a path for the enhancement of solar-energy utilization and photocatalytic performance.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (No. 21962006, 21707055, 21872030), Youth Key Project of Jiangxi Province Nature Science Foundation (20192ACBL21011), Jiangxi Province Natural Science Foundation (No. 20181BAB213010), Project Supported by Guangdong Province Universities and Colleges Pearl River Scholar Funded Scheme (2019), Key Research Project of Natural Science of Guangdong Provincial Department of Education (2019KZDXM010), Guangdong Basic and Applied Basic Research Foundation (2019A1515011249), Program of Qingjiang Excellent Young Talents, JXUST (No. 3401223429), Young Science Foundation of Jiangxi Province Education Office (GJJ180465), Open Fund of Guangdong Provincial Key Laboratory of Petrochemical Pollution Process and Control (No. 2018B030322017) and Open Fund for Key Laboratory of Green Energy and Environmental Catalysis in Universities of Fujian Province, Ningde Normal University (FJ-GEEC201901).

References

  1. H. Tong, S. Ouyang, Y. Bi, N. Umezawa, M. Oshikiri and J. Ye, Adv. Mater., 2012, 24, 229–251 CrossRef CAS.
  2. Y. Zheng, W. Wang, D. Jiang, L. Zhang, X. Lia and Z. Wang, J. Mater. Chem. A, 2016, 4, 105–112 RSC.
  3. S. B. Wang, B. Y. Guan, Y. Lu and X. W. D. Lou, J. Am. Chem. Soc., 2017, 139, 17305–17308 CrossRef CAS.
  4. N. Li, Y. Tian, J. Zhao, J. Zhang, J. Zhang, W. Zuo and Yi. Ding, Appl. Catal., B, 2017, 214, 126–136 CrossRef CAS.
  5. Y. C. Deng, L. Tang, G. M. Zeng, Z. J. Zhu, M. Yan, Y. Y. Zhou, J. J. Wang, Y. N. Liu and J. J. Wang, Appl. Catal., B, 2017, 203, 343–354 CrossRef CAS.
  6. P. Mohapatra, S. K. Samantaray and K. Parida, J. Photochem. Photobiol., A, 2005, 170, 189–194 CrossRef CAS.
  7. F. Hashemzadeh, A. Gaffarinejad and R. Rahimi, J. Hazard. Mater., 2015, 286, 64–74 CrossRef CAS.
  8. V. R. Choudhary, A. G. Gaikwad and S. D. Sansare, Angew. Chem., Int. Ed., 2001, 40, 1776–1779 CrossRef CAS.
  9. R. K. Das and A. K. Golder, J. Electroanal. Chem., 2018, 823, 9–19 CrossRef CAS.
  10. T. P. Fellinger, F. Hasché, P. Strasser and M. Antonietti, J. Am. Chem. Soc., 2012, 134, 4072–4075 CrossRef CAS.
  11. S. N. Li, G. H. Dong, R. Hailili, L. P. Yang, Y. X. Li, F. Wang, Y. B. Zeng and C. Y. Wang, Appl. Catal., B, 2016, 190, 26–35 CrossRef CAS.
  12. C. L. Yu, D. B. Zeng, Q. Z. Fan, K. Yang, J. L. Zeng, L. F. Wei, J. H. Yi and H. B. Ji, Environ. Sci.: Nano, 2020, 7, 286–303 RSC.
  13. Y. Liu, Z. Z. Xiao, S. Cao, J. H. Li and L. Y. Piao, Chin. J. Catal., 2020, 41, 219–226 CrossRef CAS.
  14. G. Zhang, G. Li, T. Heil, S. Zafeiratos, F. Lai, A. Savateev, M. Antonietti and X. Wang, Angew. Chem., 2019, 131, 3471–3475 CrossRef.
  15. C. L. Yu, D. B. Zeng, F. Y. Chen, H. B. Ji, J. L. Zeng, D. H. Li and K. Yang, Appl. Catal., A, 2019, 578, 70–82 CrossRef CAS.
  16. S. S. Xue, H. B. He, Z. Wu, C. L. Yu, Q. Z. Fan, G. M. Peng and K. Yang, J. Alloys Compd., 2017, 694, 989–997 CrossRef CAS.
  17. J. D. Hu, D. Y. Chen, N. J. Li, Q. F. Xu, H. Li, J. H. He and J. M. Lu, Appl. Catal., B, 2018, 236, 45–52 CrossRef CAS.
  18. Z. Kong, Y. J. Yuan, D. Q. Chen, G. L. Fang, Y. Yang, S. H. Yang and D. P. Cao, Dalton Trans., 2017, 46, 2072–2076 RSC.
  19. X. Zhang, J. Zhang, J. Q. Yu, Y. Zhang, Z. X. Cui, Y. Sun and B. R. Hou, Appl. Catal., B, 2018, 220, 57–66 CrossRef CAS.
  20. M. Chatti, V. N. K. B. Adusumalli, S. Gangulia and V. Mahalingam, Dalton Trans., 2016, 45, 12384–12392 RSC.
  21. Y. N. Tang, W. H. Di, X. S. Zhai, R. Y. Yang and W. P. Qin, ACS Catal., 2013, 3, 405–412 CrossRef CAS.
  22. J. T. Hou, Y. Z. Li, M. Y. Mao, Y. Z. Yue, G. N. Greaves and X. J. Zhao, Nanoscale, 2015, 7, 2633–2640 RSC.
  23. Y. H. Sang, Z. H. Zhao, M. W. Zhao and H. Liu, Adv. Mater., 2015, 27, 363–369 CrossRef CAS.
  24. H. J. Yu, R. Shi, Y. F. Zhao, G. I. N. Waterhouse, L. Z. Wu, C. H. Tung and T. R. Zhang, Adv. Mater., 2016, 28, 9454–9477 CrossRef CAS.
  25. S. W. Zeng, S. Y. Hu, J. Xia, T. Anderson, X. Q. Dinh, X. M. Meng, P. Coquet and K. T. Yong, Sens. Actuators, B, 2015, 207, 801–810 CrossRef CAS.
  26. H. F. Zhang, T. Wang, Z. X. Yang, Y. F. Liu, J. Zhao, Q. J. Li and Y. L. Mao, CrystEngComm, 2019, 21, 1019–1025 RSC.
  27. Z. B. Wu, X. Z. Yuan, G. M. Zeng, L. B. Jiang, H. Zhong, Y. C. Xie, H. Wang, X. H. Chenand and H. Wang, Appl. Catal., B, 2018, 225, 8–21 CrossRef CAS.
  28. Y. B. Liu, G. Q. Zhu, J. Z. Gao, R. L. Zhu, M. Hojamberdiev, C. H. Wang, X. M. Wei and P. Liu, Appl. Catal., B, 2017, 205, 421–432 CrossRef CAS.
  29. X. X. Li, K. Yang, C. L. Yu, S. Yang, K. L. Zhang, W. X. Dai, H. B. Ji, L. H. Zhu, W. Y. Huang and S. B. Ouyang, J. Mater. Chem. C, 2019, 7, 8053–8062 RSC.
  30. X. X. Li, K. Yang, C. L. Yu, K. L. Zhang, S. Yang, L. H. Zhu, H. B. Ji, W. X. Dai, Q. z. Fan and W. Y. Huang, Inorg. Chem. Front., 2019, 6, 3158–3167 RSC.
  31. K. L. Zhang, M. Zhou, C. L. Yu, K. Yang, X. X. Li, S. Yang, J. Guan, W. X. Dai and W. Y. Huang, J. Mater. Sci., 2020, 55, 10435–10452 CrossRef CAS.
  32. K. Fu, J. Z. Huang, N. N. Yao, X. J. Xu and M. Z. Wei, Ind. Eng. Chem. Res., 2015, 54, 659–665 CrossRef CAS.
  33. T. Liu, G. Q. Tan, C. C. Zhao, C. Xu, Y. N. Su, Y. Wang, H. J. Ren, A. Xia, D. Shao and S. M. Yan, Appl. Catal., B, 2017, 213, 87–96 CrossRef CAS.
  34. S. Ganguli, C. Hazra, M. Chatti, T. Samanta and V. Mahalingam, Langmuir, 2016, 32, 247–253 CrossRef CAS.
  35. C. Regmi, Y. K. Kshetri, S. K. Ray, R. P. Pandey and S. W. Lee, Appl. Surf. Sci., 2017, 392, 61–70 CrossRef CAS.
  36. R. Adhikari, G. Gyawali, S. H. Cho, R. Narro-García, T. Sekino and S. W. Lee, J. Solid State Chem., 2014, 209, 74–81 CrossRef CAS.
  37. G. Kresse and J. Furthmuller, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, 11169–11186 CrossRef CAS.
  38. G. Kresse and J. Furthmuller, Comput. Mater. Sci., 1996, 6, 15–50 CrossRef CAS.
  39. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868 CrossRef CAS.
  40. P. E. Blöchl, Phys. Rev. B: Condens. Matter Mater. Phys., 1994, 50, 17953–17979 CrossRef.
  41. F. Guo, W. L. Shi, Y. Cai, S. W. Shao, T. Zhang, W. S. Guan, H. Huang and Y. Liu, RSC Adv., 2016, 6, 93887–93893 RSC.
  42. J. Wang, C. H. Hua, X. L. Dong, Y. Wang and N. Zheng, Sustainable Energy Fuels, 2020, 4, 1855–1862 RSC.
  43. Q. Han, X. Bai, Z. Man and L. Li, J. Am. Chem. Soc., 2019, 141, 4209–4213 CrossRef CAS.
  44. H. Y. Hafeez, S. K. Lakhera, M. Ashokkumar and B. Neppolian, Ultrason. Sonochem., 2019, 53, 1–10 CrossRef CAS.
  45. Z. Ai, L. Zhang and S. Lee, J. Phys. Chem. C, 2010, 114, 18594–18600 CrossRef CAS.
  46. Y. X. Zhang, D. Ma, J. Wu, Q. Z. Zhang, Y. J. Xin and N. Bao, Appl. Surf. Sci., 2015, 353, 1260–1268 CrossRef CAS.
  47. J. Li, S. Tang, L. Lu and H. C. Zeng, J. Am. Chem. Soc., 2007, 129, 9401–9409 CrossRef CAS.
  48. X. Q. Zhao, H. Suo, Z. Y. Zhang, G. Zhang and C. F. Guo, Ceram. Int., 2020, 46, 3183–3189 CrossRef CAS.
  49. S. K. Ray, Y. K. Kshetri, T. Yamaguchi, T. H. Kim and S. W. Lee, J. Solid State Chem., 2019, 272, 87–95 CrossRef CAS.
  50. F. Zhang, C. L. Zhang, W. N. Wang, H. P. Cong and H. S. Qian, ChemSusChem, 2016, 9, 1449–1454 CrossRef CAS.
  51. Z. M. Fang, Q. Hong, Z. H. Zhou, S. J. Dai, W. Z. Weng and H. L. Wan, Catal. Lett., 1999, 61, 39–44 CrossRef CAS.
  52. K. Feng, S. Huang, Z. Lou, N. W. Zhu and H. P. Yuan, Dalton Trans., 2015, 44, 13681–13687 RSC.
  53. X. Chen and Z. Song, J. Opt. Soc. Am. B, 2007, 24, 965 CrossRef CAS.
  54. A. Tymiński, T. Grzyb and S. Lis, J. Am. Ceram. Soc., 2016, 99, 3300–3308 CrossRef.
  55. X. L. Pang, C. H. Jia, G. Q. Li and W. F. Zhang, Opt. Mater., 2011, 34, 234–238 CrossRef CAS.
  56. H. Wang, X. D. Hong, R. L. Han, J. H. Shi, Z. J. Liu, S. J. Liu, Y. Wang and Y. Gan, Sci. Rep., 2015, 5, 17088 CrossRef CAS.
  57. T. Li, C. F. Guo, P. J. Zhao, L. Li and J. H. Jeong, J. Am. Ceram. Soc., 2013, 1197, 1193–1197 CrossRef.
  58. C. Tong, R. Xiang, L. S. Peng, L. Q. Tan, X. Y. Tang, J. C. Wang, L. Li, Q. Liao and Z. D. Wei, J. Mater. Chem. A, 2020, 8, 3351–3356 RSC.
  59. S. P. Mohan, M. K. Purkait and C. T. Chang, Int. J. Hydrogen Energy, 2020, 45, 17174–17190 CrossRef.
  60. S. Obregón and G. Colón, Appl. Catal., B, 2014, 152–153, 328–334 CrossRef.
  61. W. K. Jo, S. Kumar, S. Eslav and S. Tonda, Appl. Catal., B, 2018, 239, 586–598 CrossRef CAS.
  62. Y. q. Zhou, E. M. Zahran, B. A. Quiroga, J. Perez, K. J. Mintz, Z. l. Peng, P. Y. Liyanage, R. R. Pandey, C. s. C. Chusuei and R. M. Leblanc, Appl. Catal., B, 2019, 248, 157–166 CrossRef CAS.
  63. G. C. Xiao, D. Z. Li, X. Z. Fu, X. X. Wang and P. Liu, Chin. J. Inorg. Chem., 2004, 20, 195–198 CAS.

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/d0dt02318c

This journal is © The Royal Society of Chemistry 2020